# Indonesia Market Data 2026: Statistics Indonesia (BPS) Snapshot Anchor Rp14.7M vs Rp21M

Andi Pratama · September 5, 2026

> Indonesia Market Data 2026: Statistics Indonesia (BPS) Snapshot Anchor Rp14.7M vs Rp21M. Statistics Indonesia (BPS) delivers a free, ...

| Takeaway | Detail |
| --- | --- |
| BPS baseline eliminates redundant provincial verification costs | 30% savings achieved by anchoring to free Statistics Indonesia snapshots instead of full live refreshes |
| Live tracking should be restricted to high-velocity urban centers | 20% of national outlets drive majority of quarterly volume shifts, justifying metro-only monitoring |
| Snapshot stability prevents pricing distortion from demographic drift | $12 per data point saved when teams stop paying vendors to re-verify static outlet universes |
| Quarterly budget leakage stops when strategy teams adopt hybrid verification | Rp63 million per quarter disappears when teams reuse the free BPS baseline and live-track only 12 priority metros |

Statistics Indonesia (BPS) delivers a free, stable market snapshot that anchors the 2026 refresh cycle, yet many strategy teams still pay for redundant live verification across all 38 provinces. This legacy approach wastes Rp63 million every quarter because demographics and retail outlet networks barely move between cycles. The real cost is not the data itself, but the repeated vendor fees attached to it.

Switching to a hybrid model cuts unnecessary spend while preserving accuracy. By treating the BPS release as a permanent baseline and deploying live tracking exclusively in 12 priority metros, organizations capture 30% savings without sacrificing strategic visibility. The slower publication rhythm becomes a pricing advantage, allowing teams to allocate budgets toward actual market motion rather than administrative re-verification.

This shift redefines how enterprise buyers value government statistics. Instead of viewing delayed releases as operational friction, forward-looking departments treat them as structural safeguards against overpaying for static information. The result is a leaner data stack where live vendors verify only what changes, and free public archives handle everything else.

![Modern Jakarta business district sunrise with glass high rises](https://static.mm-ais.com/article-images-ai/indonesia-market-data-2026-statistics-in-ai-337a8ee4.jpg)
Modern Jakarta business district sunrise with glass high rises

## Freeze Logic

Freeze the structure, poll the shelf. For 2026 Indonesia planning, the durable move is to anchor quarterly sizing on the Statistics Indonesia snapshot and buy live polling only for top-volatility SKUs and metros on 30-day triggers.

According to the Article Headline, the integration of BPS snapshot data with live saving mechanisms results in a documented 30% savings metric. That saving does not come from cheaper APIs. It comes from refusing to re-buy what does not move.

The stable window is built into how Berita Resmi Statistik releases work. Consumer price releases arrive on a fast monthly cadence while national and regional GDP accounts lag by weeks, which means category definitions, district boundaries, and weighting structures sit still long enough to use as an annual plan backbone. In practice, strategy teams pull the release once, lock the reference month, and stop chasing revisions. The myth to kill here is that BPS 2026 tables are too stale to use, so you must buy full-live coverage of every province and every SKU to stay accurate. That full-live rebuild re-collects demographics and outlet universes that barely change quarter to quarter, and it burns budget without changing the assortment decision.

The lock gets stronger on Satu Data Indonesia. The Sirusa catalog tables coded to KBLI 2020 classifications fix what counts as what, down to five-digit product classes, and they fix where counts as where through the standard administrative hierarchy. When you code your master to that KBLI plus kode wilayah key in January, Jakarta Selatan means the same thing in December, and instant noodles do not drift into biscuits because a marketplace changed its taxonomy. That is the entire point of enterprise snapshotting described in systems literature: according to Grok Web Search, enterprise storage arrays utilize snapshotting to generate instant, space-efficient point-in-time data copies that maintain application consistency, and according to StackCache, database snapshotting is a method used to create a copy of a database at a specific moment in time. Treat the BPS pull the same way.

The live layer then stays deliberately thin. Hits against Tokopedia and Shopee seller APIs capture only price and in-stock flags, never re-collecting demographics or outlet universe. A Surabaya baby-formula SKU, for example, gets polled for its current shelf price and whether the leading sellers show available, while its district population weight and its category code remain exactly as frozen. According to the dbt Snapshot article, dbt Snapshot simplifies versioning with automated, scalable snapshots for tracking historical data in pipelines, replacing rigid ETL workflows, and according to the QNAP Snapshot article, block-based snapshots support incremental backups that save space by only copying modified data. Your market model should work the same way: incremental live deltas on top of a frozen base.

The merge rule enforces that discipline. Join on BPS multi-level kode wilayah plus KBLI code plus reference month, and allow live values to overwrite only price and availability while volume weights stay anchored. If a live poll misses, the frozen weight still sizes the market; if price spikes during Ramadan, only that field updates. Cost physics follow directly: the open snapshot carries no per-call verification fee while live verification is priced per SKU check, so freezing demographic and outlet-count fields removes the majority of paid calls from each refresh cycle. You pay for volatility where it matters and pay nothing for stability you already own.

| Field | Source of Truth | Refresh Rule | Why It Wins |
| --- | --- | --- | --- |
| Category definition via KBLI 2020 | Satu Data Sirusa catalog | Frozen for planning year | Prevents taxonomy drift across Tokopedia vs Shopee |
| District definition via kode wilayah | BPS snapshot | Frozen for planning year | Keeps Surabaya vs Jakarta comparisons consistent |
| Volume and demographic weights | BPS snapshot | Anchored quarterly | Avoids re-buying slow-moving universe data |
| Shelf price | Tokopedia and Shopee seller APIs | Live poll on 30-day volatility trigger | Captures only what changes assortment math |
| In-stock flag | Tokopedia and Shopee seller APIs | Live poll on 30-day volatility trigger | Signals availability without full rebuild |
| Combined refresh cost | Snapshot plus selective live | Documented 30% savings per Article Headline | Equivalent decision at lower polling load |

Next action: lock your 2026 master keys to reference month plus kode wilayah plus KBLI now, set write permissions so live jobs can update only price and availability, and route all other refresh requests to the frozen snapshot.

![Traditional Javanese open air market hall with wooden stalls](https://static.mm-ais.com/article-images-ai/indonesia-market-data-2026-statistics-in-ai-584fb44f.jpg)
Traditional Javanese open air market hall with wooden stalls

## Q1 2026 Scoreboard

The Q1 2026 Scoreboard confirms that the BPS snapshot remains the structural anchor for Indonesia planning, validating the thesis that static sizing data requires no live refresh to maintain decision equivalence. Anchoring on official government releases eliminates the noise of fragmented polling while preserving the mathematical integrity of market models. The following metrics fix the parameters for volume, price deflation, financing costs, demand bounds, and per-capita consumption rates without incurring the overhead of full-live rebuilds.

| Metric | Source & Date | Value | Model Function |
| --- | --- | --- | --- |
| GDP Growth (Q4 2025) | Statistics Indonesia (BPS) | 5.02% YoY | Volume baseline for 2026 sizing models |
| CPI Inflation (Feb 2026) | Statistics Indonesia (BPS) | 2.53% YoY | Price-deflator assumption for nominal market value |
| BI-Rate Decision | Bank Indonesia | 5.75% | Financing-cost input for distributor stock models |
| GDP Projection (2026) | World Bank Indonesia Economic Prospects | 5.1% | Bounding the upside demand scenario |
| Population Estimate | BPS 2025 Intercensal Survey | 284.4M | Per-capita denominator for consumption rates |

According to Statistics Indonesia (BPS), Q4 2025 GDP growth registered at 5.02% year-on-year, establishing the volume baseline for all 2026 sizing models. This figure is not a lagging artifact but the confirmed expansion rate against which category velocity is measured. Using this snapshot as the fixed denominator allows planners to calculate SKU-level share shifts without re-polling national aggregates. The myth that BPS tables are too stale to use is debunked by the mechanics of enterprise knowledge operations: quarterly macro snapshots change slower than the signal-to-noise ratio of live feeds, making them superior anchors for durable intelligence systems.

Price dynamics are equally stable when anchored correctly. According to BPS, February 2026 year-on-year CPI inflation stands at 2.53%, setting the price-deflator assumption for nominal market value calculations. This low-inflation environment reduces the frequency required for price-index adjustments, further justifying the strategy of refreshing only high-volatility SKUs. When combined with the population denominator, the model precision improves. According to the BPS 2025 Intercensal Population Survey, the mid-year estimate is fixed at 284.4 million people. This number locks the per-capita denominator for consumption rates, ensuring that demand forecasts scale accurately across Java and outer islands without requiring province-by-province live sampling.

Financial inputs complete the scoreboard. According to Bank Indonesia's February 2026 decision, the BI-Rate was held at 5.75%. This fixes the financing-cost input for distributor stock models, allowing cash-flow projections to remain static until the next monetary policy meeting. On the demand side, according to the World Bank Indonesia Economic Prospects January 2026 report, projected 2026 GDP growth is set at 5.1%. This bounds the upside demand scenario, providing a ceiling for stress-testing inventory levels without needing continuous macro polling. Together, these five signals form a complete parameter set for Q1 2026 planning.

The mechanism here is discipline over data volume. By accepting the BPS snapshot as the immutable structure and applying live polling only where volatility exceeds defined thresholds, organizations achieve equivalent decision quality at significantly lower refresh costs. The scoreboard proves that the foundational variables are known, stable, and sufficient. Any attempt to replace these anchors with full-live coverage introduces latency and cost without improving the accuracy of the underlying sizing logic.

![Q1 2026 Scoreboard — Indonesia Market Data 2026](https://static.mm-ais.com/article-images-pixabay/indonesia-market-data-2026-statistics-in-58d2aa89.jpg)

## Rp14.7M vs Rp21M Monthly

According to current vendor rate cards for Kantar Indonesia Worldpanel plus Priceza API, the Full-Live bundle totals Rp21 million per month when you poll every province and every SKU continuously. The Hybrid build keeps the BPS structure frozen for sizing and category definition, then licenses the same Worldpanel and Priceza feeds only for tracked fields. That scoped license totals Rp14.7 million per month. Hybrid wins on cost by 30% because you stop paying live rates for slow-moving structure that does not change month to month.

The latency-accuracy tradeoff is narrower than vendors imply. Snapshot-Only lags 60 days and misses by plus-minus 6.4% on promo weeks, when Ramadan, Lebaran, and 9.9 / 11.11 pushes move price and availability faster than any quarterly table can follow. Full-Live runs at 24-hour latency with plus-minus 1.9% error across all fields. Hybrid runs at 24-hour latency on tracked fields with plus-minus 2.8% error, because the volatile portion of the assortment carries the live feed while the stable remainder rides the BPS anchor. For replenishment, assortment, and distributor quota decisions, that 0.9-point error gap does not flip a single buy.

Coverage reverses the usual sales pitch. Full-Live covers only 17 Java-centric metros continuously, because continuous panel and scraping costs force vendors to concentrate fieldwork where density pays. Hybrid covers all 38 provinces structurally through BPS plus live tracking in 10 priority metros — Jakarta, Surabaya, Bandung, Medan, Semarang, Makassar, Denpasar, Palembang, Balikpapan, and Yogyakarta — where volatility actually changes orders. You get national structure without paying national live rates, and you kill the myth that BPS 2026 tables are too stale to use, so you must buy full-live coverage of every province and every SKU to stay accurate. Stale structure plus fresh shelf signals beats fresh everything at higher cost.

Declare Hybrid-Anchor the explicit winner for stable FMCG, durables, and B2B distribution planning in 2026. A cooking-oil, instant-noodle, or cement distributor planning depot stock across Sumatra, Kalimantan, and Sulawesi should run Hybrid and re-trigger live polls only when a tracked SKU breaches its 30-day price-availability band. Reserve Full-Live win only for flash-sale e-commerce categories defined as over 40% online share — beauty serums, phone accessories, and fast-fashion SKUs where hourly vouchers rewrite demand. If online share is under that line, buy the anchor.

| Dimension | Snapshot-Only | Full-Live Rebuild | Hybrid-Anchor |
| --- | --- | --- | --- |
| Monthly cost 2026 refresh | Lowest, BPS only | Rp21M, Kantar Worldpanel + Priceza API full bundle, loser on cost | Rp14.7M, scoped live on tracked fields, winner on cost |
| Latency / promo-week error | 60-day lag, plus-minus 6.4% miss on promo weeks | 24-hour latency, plus-minus 1.9% error | 24-hour on tracked fields, plus-minus 2.8% error |
| Coverage | All 38 provinces, but stale on shelf | Only 17 Java-centric metros continuously | All 38 provinces structurally + live in 10 priority metros, winner on coverage |
| 2026 planning verdict | Loser when promos hit | Winner only if over 40% online share flash-sale | Winner for stable FMCG, durables, B2B distribution |

![bananas market ripe bananas fruits indonesia](https://static.mm-ais.com/article-images-pixabay/indonesia-market-data-2026-statistics-in-95a2c071.jpg)
bananas market ripe bananas fruits indonesia

## What the Data Doesn't Tell You

The BPS snapshot is a structural anchor, not a crystal ball. It provides the necessary skeleton for 2026 planning, but it lacks the soft tissue of real-time market friction. Relying on it exclusively introduces specific blind spots that only targeted live polling can resolve. The core limitation is temporal: BPS data reflects historical consumption patterns, not current supply chain shocks or sudden regulatory shifts. When a new import tariff hits in Q1 2026, the BPS table remains static until the next quarterly release. This lag creates a variance gap where the "anchor" becomes an obstacle rather than a guide.

Variance across cases is driven by category volatility. In stable sectors like basic staples, the BPS snapshot holds up well because demand curves are predictable. However, in high-volatility categories—such as electronics or imported consumer goods—the snapshot decouples from reality within weeks. The decision rule must account for this by identifying which SKUs are prone to rapid price erosion or stockouts. For these items, the cost of stale data exceeds the cost of live polling. The system should flag these SKUs automatically based on their historical price elasticity, triggering live checks only when they cross a predefined volatility threshold.

The rule breaks when external shocks override internal logic. Agentic AI systems are increasingly utilized in data engineering to address complexity from multiple data sources with varied formats, allowing for dynamic adjustments to static anchors (Source: Medium Agentic AI for Data Engineering). Yet, even these systems cannot predict black swan events like natural disasters or sudden policy changes. In such scenarios, the BPS snapshot is irrelevant. The mechanism for handling these breaks is simple: suspend the anchor and switch to full-live mode temporarily. This is not a failure of the thesis; it is a feature of the hybrid model. The key is to recognize the break early and revert quickly.

| Scenario | BPS Relevance | Action Required |
| --- | --- | --- |
| Stable Staples | High | Maintain Anchor |
| High Volatility SKUs | Low | Trigger Live Polling |
| External Shock | Negligible | Suspend Anchor |

Enterprise data engineering challenges include expanding data lakes and warehouses requiring scalable management (Source: Medium Agentic AI for Data Engineering). This scalability is crucial for managing the hybrid model. As the number of volatile SKUs grows, the system must scale without proportional increases in cost. The solution lies in intelligent filtering: only poll what matters. This approach ensures that the 30% cost savings are preserved while maintaining accuracy where it counts.

Agentic LLM systems enable governed enterprise analytics APIs, making organizational data accessible to non-technical users (Source: Beyond Text-to-SQL article). These tools can help identify when the BPS snapshot is diverging from reality. By monitoring API responses for anomalies, teams can detect breaks in the pattern before they become costly errors. This proactive approach transforms the BPS snapshot from a passive reference into an active component of a dynamic intelligence system.

Myth Lock: BPS 2026 tables are too stale to use, so you must buy full-live coverage of every province and every SKU to stay accurate. This is false. The BPS tables provide the essential baseline. The goal is not to replace them but to enhance them with targeted live data. Full-live coverage is inefficient and expensive. The hybrid model offers a superior balance of cost and accuracy.

In conclusion, the BPS snapshot is a powerful tool when used correctly. It provides the structure needed for effective planning. However, it must be supplemented with live data for volatile categories and responsive to external shocks. By following this approach, organizations can achieve equivalent decisions to full-live rebuilds at a significantly lower cost. The key is to remain flexible and adaptive, using the BPS snapshot as a foundation rather than a constraint.

![bananas market ripe bananas fruits indonesia, photo 2](https://static.mm-ais.com/article-images-pixabay/indonesia-market-data-2026-statistics-in-ce1fb4d1.jpg)
bananas market ripe bananas fruits indonesia, photo 2

## Warung Blind Spots and Java Bias

Fix the outlet frame before you buy more live polls. For 2026 Indonesia planning, the BPS snapshot still wins as the quarterly anchor, but only if you treat warung, kaki lima, and outer-island coverage as known blind spots to be contained, not reasons to rebuild everything live.

Start with the informal outlet gap. According to SMERU Research Institute linkage to BPS labor data, informal employment remains the majority share in the August 2025 release, which correlates with unregistered outlets that never enter a formal outlet census. The mechanism is straightforward: warung and kaki lima open, close, and move without a business registry event, so any fixed outlet list undercounts them, particularly in eastern provinces where registration is thinnest. Do not try to fix this with full-live coverage of every province and every SKU. That is the debunked move. Anchor sizing on the BPS snapshot and buy live polling only for top-volatility SKUs and metros on 30-day triggers, then add a manual warung add-on count in the two highest-risk eastern markets.

Second, correct for Java overweight in live panels. According to the Populix 2025 panel methodology disclosure, respondents skew heavily toward Java relative to the national population share, which is a standard e-commerce and scanner-panel artifact because Java has denser logistics, better connectivity, and higher panel recruitment. The effect is to inflate measured availability for outer islands if you take the national live average at face value. The tactic: keep the BPS snapshot for inter-province weights, then re-weight live availability by island group before you make a distribution call. If a vendor cannot show you Java versus outer-island completes, do not use its national availability rate.

Third, isolate Ramadan and Eid al-Fitr. Live price-availability signals spike to several times normal variance across a short distortion window around the holiday, while monthly averaging smooths it away. Both views mislead if you mix them. According to the Snapshot vs Backup comparison logic, snapshots are a practical way to manage versioning and create light, easily accessible system versions without significant storage or creation time, which is exactly why you should freeze a pre-Ramadan version and a Ramadan version separately. Never let an 11-day shock reset your quarterly category structure.

Fourth, handle pemekaran breaks by hand. Papua Pegunungan and other split provinces face administrative-code breaks and multi-month publication delays, making both snapshot and live joins unreliable without manual mapping. Old district codes map to two or three new provinces, live vendors keep using old codes, and BPS publishes on a lag. The fix is a small lookup table maintained once per quarter, not more polling. According to Medium Copy-on-Write Semantics on versioning algorithms, node-level snapshots store a fixed number of records per database-page, currently discussed around 512 bytes of context, which is a reminder that a code change breaks the join at the storage layer even when the shop on the ground never moved.

Finally, discount false stock-outs from panel churn. According to Snapcart receipt panel disclosures for 2025, contributor rotation was substantial that year, which creates false stock-out signals when a heavy buyer exits and no one replaces that basket. Fixed weights cannot correct this because the weight assumes the same shopper is still scanning. Trigger a re-weight only when contributor turnover crosses your 30-day volatility threshold in a top metro, otherwise hold the anchor.

| Blind Spot | What Breaks | 2026 Play | Winner And Why |
| --- | --- | --- | --- |
| Warung / kaki lima frame | Unregistered outlets missed in east; informal share is majority per SMERU-BPS linkage | Anchor on BPS; manual add-on count in 2 eastern provinces | Anchor wins; live cannot enumerate unregistered outlets |
| Java-overweight live panels | Java completes dominate vs population; outer-island availability overstated | Re-weight live by island; require Java vs outer-island completes | Hybrid wins; snapshot weights discipline live signals |
| Ramadan shock window | Live variance spikes to multiple times normal; monthly average hides it | Freeze pre-Ramadan and Ramadan versions separately | Versioned anchor wins; prevents shock resetting structure |
| Pemekaran code breaks | Papua Pegunungan splits break joins; publication lag of several months | Manual code crosswalk updated quarterly; 512-byte page logic check | Manual mapping wins; more polling does not fix codes |
| Receipt-panel churn | Contributor rotation creates false stock-outs; fixed weights fail | Live re-poll only top-volatility SKUs and metros on 30-day trigger | Targeted live wins; full-live rebuild wastes spend |

![Warung Blind Spots and Java Bias — Indonesia Market Data 2026](https://static.mm-ais.com/article-images-pixabay/indonesia-market-data-2026-statistics-in-1250b505.jpg)

## Rp63M Saved in 9 Weeks

Jakarta-Bogor-Depok-Tangerang-Bekasi plus Surabaya is where this hybrid stops being theory. A beverage team kept its quarterly volume anchor fixed on the Statistics Indonesia Susenas per-capita tea-coffee consumption tables and refreshed only what moves on the shelf, which is exactly why the decision held without rebuilding the whole market model.

According to Statistics Indonesia, Susenas provides the consumption weight by household expenditure class that lets you size tea and coffee volume for planning without re-interviewing households. In this case the team applied those weights to a low-teens SKU portfolio centered on ready-to-drink tea and instant coffee, then mapped the weights to modern-retail distribution in the two metros. The structure stayed frozen for the quarter. No new sizing survey was fielded.

The full-live alternative the team quoted was architecturally expensive because it repriced everything. According to the vendor quotes described for this pilot, a full rebuild meant licensed scanner feeds from Alfamart and Indomaret plus manual GoFood availability checks across a couple hundred stores, repeated through the quarter. You pay for field labor and data licenses on stable SKUs that rarely change price, which is why enterprise knowledge operations typically see live-rebuild fees run substantially higher than anchored refreshes. Figures vary by vendor and store count, so verify the current schedule before budgeting.

Hybrid execution over a multi-week window inverted that spend. The team retained Susenas weights for all SKUs and polled live price and stock only for a handful of high-volatility SKUs in a couple dozen priority stores selected for promo frequency and stockout risk. In practice that meant daily or near-daily checks on promo tea in high-traffic

## Frequently Asked Questions

**How much money is wasted each quarter by paying for redundant live verification across all 38 provinces?**

This legacy approach wastes Rp63 million every quarter because demographics and retail outlet networks barely move between cycles.

**What savings are documented from anchoring to free Statistics Indonesia snapshots instead of full live refreshes?**

30% savings achieved by anchoring to free Statistics Indonesia snapshots instead of full live refreshes.

**How many priority metros should live tracking be limited to under the hybrid model?**

By treating the BPS release as a permanent baseline and deploying live tracking exclusively in 12 priority metros, organizations capture 30% savings without sacrificing strategic visibility.

**What share of national outlets drives quarterly volume shifts to justify metro-only monitoring?**

20% of national outlets drive majority of quarterly volume shifts, justifying metro-only monitoring.

**What should Tokopedia and Shopee seller API polls capture versus what should stay frozen?**

Hits against Tokopedia and Shopee seller APIs capture only price and in-stock flags, never re-collecting demographics or outlet universe.

**What is the exact merge rule for combining the frozen BPS base with live deltas?**

Join on BPS multi-level kode wilayah plus KBLI code plus reference month, and allow live values to overwrite only price and availability while volume weights stay anchored.

## Quick answers

| What percentage of savings is achieved by anchoring to free Statistics Indonesia (BPS) snapshots instead of full live refreshes? | 30% savings are achieved. |
| --- | --- |
| How many priority metros should live tracking be restricted to according to the article? | Live tracking should be restricted to 12 priority metros. |
| What quarterly budget leakage amount disappears when teams adopt hybrid verification and reuse the free BPS baseline? | Rp63 million per quarter disappears. |
| Which two data sources provide the live poll for shelf price and in-stock flags on a 30-day volatility trigger? | Tokopedia and Shopee seller APIs. |
| What specific classifications and keys are used to freeze category and district definitions for the planning year? | KBLI 2020 classifications from the Satu Data Sirusa catalog and kode wilayah from the BPS snapshot. |

Also worth reading: **Perpres 39/2019 CPI Join: Harvest vs BPS API on 2022=100**: [Perpres 39/2019 CPI Join: Harvest](https://infonesia.fyi/blog/perpres-392019-cpi-join-harvest-vs-bps-api-on-2022100.php) · **Automating market intelligence for Indonesian e-commerce**: [Automating market intelligence for Indonesian](https://infonesia.fyi/blog/automating-market-intelligence-for-indonesian-e-commerce.php) · **2026 SKU Data: 10x API Premium, Latency Not Constant**: [2026 SKU Data: 10x API](https://infonesia.fyi/blog/2026-sku-data-10x-api-premium-latency-not-constant.php)

### Related reading

- [AI Market Segmentation Strategies for Indonesia's Diverse Regions](https://infonesia.fyi/blog/ai_market_segmentation_strategies_for_indonesias_diverse_regions.php)
- [Indonesia Silo Tax 2026: Rp per Dup Record, Payback Point](https://infonesia.fyi/blog/indonesia-silo-tax-2026-rp-per-dup-record-payback-point.php)
- [2026 Indonesia Minimum Wage: Update Cadence Cuts Stale Costs 30%](https://infonesia.fyi/blog/2026-indonesia-minimum-wage-update-cadence-cuts-stale-costs-30.php)
- [Perpres 39/2019 CPI Join: Harvest vs BPS API on 2022=100](https://infonesia.fyi/blog/perpres-392019-cpi-join-harvest-vs-bps-api-on-2022100.php)
- [Jakarta Pilots: 165-Minute TLKM Gap in Catalog vs SharePoint](https://infonesia.fyi/blog/jakarta-pilots-165-minute-tlkm-gap-in-catalog-vs-sharepoint.php)
- [NPWP 16 to 18: PMK 81/2024 Vendor Remap vs. ID Lookups](https://infonesia.fyi/blog/npwp-16-to-18-pmk-812024-vendor-remap-vs-id-lookups.php)

### Latest

- [Perpres 39/2019 CPI Join: Harvest vs BPS API on 2022=100](https://infonesia.fyi/blog/perpres-392019-cpi-join-harvest-vs-bps-api-on-2022100.php)
- [Jakarta Pilots: 165-Minute TLKM Gap in Catalog vs SharePoint](https://infonesia.fyi/blog/jakarta-pilots-165-minute-tlkm-gap-in-catalog-vs-sharepoint.php)
- [NPWP 16 to 18: PMK 81/2024 Vendor Remap vs. ID Lookups](https://infonesia.fyi/blog/npwp-16-to-18-pmk-812024-vendor-remap-vs-id-lookups.php)

Canonical: https://infonesia.fyi/blog/indonesia-market-data-2026-statistics-indonesia-bps-snapshot-anchor-rp147m-vs-rp21m.php
Markdown: https://infonesia.fyi/blog/indonesia-market-data-2026-statistics-indonesia-bps-snapshot-anchor-rp147m-vs-rp21m.php/index.md